Sakana Replicates Picbreeder Open-Ended Evolution
kaixhin · x · 2026-07-10
Sakana AI released an experiment on Picbreeder: using Vision-Language Model (VLM) agents to replicate an open-ended evolutionary process with "no preset goals, relying solely on humans selecting interesting images."
They revisited Picbreeder, an early collaborative creation website where users continuously selected "interesting-looking" images across multiple generations, eventually breeding unexpected forms like faces, animals, vehicles, and skulls. The team re-implemented this process as a VLM agent system, allowing agents to autonomously explore the search space. They discuss the necessary ingredients for open-ended creativity in their #GECCO2026 paper, "In Search of the Ingredients of Open-Endedness: Replicating Picbreeder with Large Vision-Language Models."
Related event: Sakana AI Replicates Picbreeder with VLM Agents for Open-Ended Creativity(14 posts)→
More from Multimodal
- Reddit user chains Ideogram 4 and Krea2 to mimic bbox-based image positioning — v3lh0t05c0 · 2026-07-22
- Ultimate Face Fix: Open-Source Multi-Face Repair Node for ComfyUI — Merserk13 · 2026-07-22
- Getting Started with AI Video: Solving Consistency and Censorship — cynicalnewenglander · 2026-07-22
- Storyboard-first workflows are making AI dance videos and influencers more consistent — aftahi_ai · 2026-07-22
- Runpod MCP and Claude help spin up image and video generation workflows — 802high · 2026-07-22
- Midjourney prompt turns a bee into a glitching pixel explosion — michaelrabone · 2026-07-22